C 2005)
Journal of Chemical Ecology, Vol. 31, No. 1, January 2005 (
DOI: 10.1007/s10886-005-0980-6
GENDER DIMORPHISM AND ALTITUDINAL VARIATION
OF SECONDARY COMPOUNDS IN LEAVES OF THE
GYNODIOECIOUS SHRUB Daphne laureola
CONCHITA ALONSO,1,∗ RICARDO PÉREZ,2 PEDRO M. NIETO,2
and JULIO DELGADO2
1 Estación
Biológica de Doñana, Consejo Superior de Investigaciones
Cientı́ficas, Apdo. 1056, E-41080 Sevilla, Spain
2 Instituto
de Investigaciones Quı́micas, Centro de Investigaciones Cientı́ficas
Isla de La Cartuja, Consejo Superior de Investigaciones Cientı́ficas
Avda. Américo Vespucio s/n, E-41092 Sevilla, Spain
(Received February 5, 2004; accepted August 31, 2004)
Abstract—In this article, we analyzed the concentration of coumarins in leaves
of female and hermaphrodite individuals of the gynodioecious shrub Daphne
laureola, along an elevational gradient in southern Spain. Combining HPLC
and NMR techniques, we identified three different glycosides of 7-methoxycoumarin in leaves of this species. Total coumarin concentration averaged between 60 and 120 mg/g dry weight for mature summer leaves of D. laureola
growing at six different populations. As predicted by optimal theory, females
tended to have a higher concentration of coumarins than hermaphrodites, thus
upholding the idea that male reproductive function is costly for hermaphrodites.
Furthermore, concentrations in females but not hermaphrodites were positively
correlated with increasing population altitude, and the magnitude of gender divergence in coumarin concentration varied among populations, suggesting that
the cost of the male function may be context dependent. To our knowledge, this is
the first evidence of gender differences in chemical defenses of a gynodioecious
species in the field.
Key Words—Coumarins, Daphne laureola, elevation, gynodioecy, plant–animal
interactions, plant defense.
∗
To whom correspondence should be addressed. E-mail: [email protected]
139
C 2005 Springer Science+Business Media, Inc.
0098-0331/05/0100-0139/0 140
ALONSO ET AL.
INTRODUCTION
In this study, we analyze the relationship between allocation to reproduction and
chemical defenses, and its natural variation along an altitudinal gradient, in a longlived perennial shrub. According to optimal theory, organisms allocate resources
to survival, growth, and reproduction, such that their fitness will be maximized
(Maynard-Smith, 1978). Plant defenses aim to decrease pathogen and herbivore
attacks that may eventually reduce plant survival, growth, and reproduction and,
thus, plant allocation to defenses can also be interpreted within optimal theory
(Herms and Mattson, 1992). Within this framework, trade-offs (or opportunity
costs) between growth, defense, and reproduction should exist whenever resources
are limited. However, field measurements of trade-offs and other physiological
costs in plant allocation to reproduction and defenses are not straightforward (see
Obeso, 2002; Cipollini et al., 2003 for a review of each topic, respectively). Gender
dimorphic species split the range of individual variance in reproductive allocation into categories, thus facilitating the assessment of consequences in terms of
survival, growth, and defense of a higher reproductive allocation under natural
conditions.
In dioecious species, allocation to reproduction in males is commonly higher
than in females at flowering, though the opposite is true at fruiting time when
the sink of resources to reproduction is maximum (Delph, 1990). Therefore, both
the nature and magnitude of differences between genders in resource allocation
may change with time (Ågren et al., 1999). Overall, herbivores usually distinguish
and prefer males over females of dioecious plant species (Ågren et al., 1999, and
references therein) suggesting that differences in reproductive allocation modify
other plant features that in turn decrease quality of females as food for herbivores
(e.g., Boecklen et al., 1990; Jing and Coley, 1990). Gynodioecy consists of populations having both hermaphrodite and female pollenless individuals. Thus, gender
difference in allocation to reproduction is in principle lower and due to the male
function only. The costs of male function are apparently more subtle, but still relevant (Eckhart and Seger, 1999), and there is also evidence for male-biased flower
and seed predation favoring the maintenance of females in gynodioecious populations (Uno, 1982; Marshall and Ganders, 2001; Collin et al., 2002; reviewed in
Ashman, 2002). To our knowledge there are no data available comparing physical
or chemical defenses in gynodioecious species (but see Gouyon and Vernet, 1980)
and, thus, whether gender differential consumption is mediated through distinct
defenses or otherwise, for example through differential attractive properties, is
unknown.
We explore the relationship between allocation to reproduction and defense
in a gynodioecious species. Hermaphrodite and female Daphne laureola plants
produce a similar number of flowers and fruits, and fruit size is also similar between genders (Alonso and Herrera, 2001). However, female flowers are smaller
GENDER DIMORPHISM AND ALTITUDINAL VARIATION OF SECONDARY COMPOUNDS
141
and pollenless, thus leading to a lower allocation to reproduction of female individuals. In southern Spain, D. laureola is consumed by several species of Noctuid
caterpillars. Defoliation has been related to plant architecture (Alonso and Herrera,
1996) and leaf nutrient composition (Alonso and Herrera, 2003), and caterpillars
are able to distinguish among several plant structures (Alonso and Herrera, 2000).
The role of allelochemicals in this plant-herbivore interaction remained unstudied. Coumarins are effective chemical defenses against herbivores in other plant
species (Berenbaum, 2001 and references therein), and were known to exist in the
Daphne genus (Hegnauer, 1973; Ulubelen et al., 1986; Zobel and Brown, 1988).
Our expectation was that, if the increased reproductive allocation associated with
the male function were costly, D. laureola females should have a higher concentration of coumarins than hermaphrodite conspecifics. Thus, we evaluated the
concentration of the three most abundant coumarin glycosides found in leaves of
female and hermaphrodite D. laureola individuals.
Furthermore, since the natural concentration of chemical defenses may vary
geographically (Johnson and Scriber, 1994), the study was conducted in six different populations selected along an altitudinal gradient, that ranged from 950 to
1800 m asl. Increased exposure to UV radiation and low temperatures at higher
altitudes may select for different chemical profiles in plants that in turn can affect
herbivores (Johnson and Scriber, 1994; Stratmann, 2003). In particular, contents
of UV-B absorbing compounds tend to increase at higher elevation sites in several
plant species (Rozema et al., 1997). Coumarins are able to absorb UV radiation
(Murray et al., 1982). Thus, we also expected an increase of coumarin concentrations in plants at higher elevation sites.
METHODS AND MATERIALS
Plant Species and Study Area. Daphne laureola L. (Thymelaeaceae) is a
long-lived evergreen shrub distributed throughout the Palearctic region and generally found in the understory of coniferous and mixed montane forests in the
Mediterranean area. In the Natural Park of Sierras de Cazorla, Segura y Las Villas
(Jaén province, south-eastern Spain), where this study was conducted, the species
is gynodioecious, and the proportion of female plants varies with site altitude
(Alonso and Herrera, 2001).
In June 2002, we collected undamaged leaves from female and hermaphrodite
individuals in six different populations comprising the entire altitudinal range of the
species at our study area. Study locations were Coto del Valle (950 m elevation),
Roblehondo (1235 m), Cañada del Espino (1575 m), Nava de las Correhuelas
(1615 m), Cabeza del Tejo (1640 m), and Puerto Llano (1800 m), hereafter referred
to as CV, RH, CÑE, NC, CT, and PLL, respectively. Aiming to have three replicates
for each combination of gender per population, leaves of each individual (N = 7–
20) were collected independently and later split into three different sets. Each
142
ALONSO ET AL.
replicate had 8–17 g fresh mass. Leaves were collected when plants were bearing
mature fruits and all plants were at the same phenological stage despite the fact
that the different populations were collected on different dates due to altitudinal
variation in plant phenology.
Chemical Analyses. Leaves were washed, dried with filter paper, and stored
at −80◦ C prior to analysis. Frozen samples were weighed, deep-frozen with liquid
nitrogen and ground in a coffee mill. As an internal control, 1.5 mg of esculetin
(6,7 dihydroxy-coumarin, Aldrich) were added to each sample to evaluate potential process errors. Leaf powder was extracted 2 × 24 hr with methanol (80%),
and the combined extract was filtered and concentrated to dryness under reduced
pressure. The residue was dissolved in water and cleaned by sequential decantation with chloroform. Coumarins were detected as pale-blue spots on C18 -TLC
(Alugram RP-18W/UV254 ) only in the aqueous phase, that was subsequently
concentrated to dryness under reduced pressure to record the weight of the final
residue. The residue was suspended in 15 mg of double-distilled water to obtain
a 10% concentration of the internal control, esculetin. Three 750 µl aliquots were
taken from this solution and centrifuged for 10 min at 12,000 rpm. The supernatant contained the target coumarins. For every aliquot, solid phase extraction
was conducted on packed MFE C18 3/500 columns (Análisis Vı́nicos S.L.) prior to
HPLC analysis. On each occasion, a new packed column was gently washed with
distilled water. The sample supernatant (250 µl) was loaded and elution started
with 750 µl of methanol (85%) that was discarded. A mixture of methanol–water–
glacial acetic acid (60:40:1, v/v; 1500 µl) was collected directly into an HPLC
vial. Two injections of 10 µl from each vial were analyzed by HPLC, and the
average peak area for each compound was used for further statistical analyses
(see below).
HPLC was conducted on a Waters 2690 separation module with a Waters
996 PDA detector (Waters Cromatografı́a S. A., Barcelona, Spain) that allowed
coumarin identification by way of their characteristic UV spectrum with two maxima absorption lengths around 260 and 320 nm (Murray et al., 1982). A Waters
ODS2-3 µm RP-HPLC column (4.6 mm i.d. × 15 cm length) was used for quantification. Analyses were conducted in isocratic mode at a flow-rate of 1 ml/min,
using a mixture of water–methanol–glacial acetic acid (84.8:14.2:1, v/v) as mobile
phase (modified from Thompson and Brown, 1984). Double distilled water and
HPLC quality solvents were used for the analyses. A calibration regression line
was obtained for esculetin by varying the volume injected of two different solutions w/w in methanol. Regression of peak area on amount of esculetin injected
explained 99.8% of peak area variation. Analyses of esculetin recovery based on
this calibration showed that on average 76% of the esculetin initially added to leaf
samples was lost during sample processing. Thus, error for each individual sample
was calculated as the ratio between the expected and observed area of esculetin
peak.
GENDER DIMORPHISM AND ALTITUDINAL VARIATION OF SECONDARY COMPOUNDS
143
Each coumarin was purified by liquid chromatography, and the 1 H and 13 C
NMR (DMSO-d6 ) obtained on a Bruker AVANCE 500 spectrometer were compared to literature data for molecule identification.
Data Analyses. Statistical analyses were performed using the SAS statistical
package (SAS Institute, 1996). Peak areas were transformed into coumarin concentrations assuming that for each sample coumarin quantification had the same
recovery error as that observed for esculetin, and using esculetin regression to transform peak areas into quantities. Results obtained for each aliquot were averaged
by sample, and concentration was referred to total leaf dry weight of the sample.
Differences between genders and populations on concentration of coumarins were
analyzed by General Linear Models (Procedure GLM). Gender, population, and
their interaction were treated as fixed effects.
RESULTS
The methanolic extract of D. laureola leaves contained three major components (Figure 1) that were identified as three different glycosides of 7-methoxycoumarin. The observed molecular structures based on 1 H and 13 C NMR, and the
references where these compounds were previously reported (Konishi et al., 1993;
FIG. 1. HPLC chromatogram of the methanolic extract of D. laureola leaves with esculetin
added, recorded at 320 nm wave length. Key to peak identity: Peak 1: 5-O-β−D-glucosyl7-methoxy-8-hydroxy coumarin (1); Peak 2: 5-O-β−D-glucosyl-(6←1)-β-glucosyl)-7methoxy-8-hydroxy coumarin (2); Peak 3: esculetin; Peak 4: 5-hydroxy-7-methoxy-8-Oβ−D-glucosyl coumarin (3).
144
ALONSO ET AL.
Jung et al., 1994) are shown in Table 1. The major component (1) averaged 54,938
(±17,646) ppm leaf dry weight, whereas average concentration of the other two
compounds were 10,403 (±4182) ppm and 17,940 (±5787) ppm for (2) and (3),
respectively. Concentrations of all three D. laureola coumarins into a sample were
positively correlated (N = 42; 0.78 < r < 0.91; P < 0.001 in all cases).
Populations differed in the concentration of the major coumarin glycoside
(1) (F5,30 = 4.56, P = 0.003; Figure 2). Population × gender interaction was not
statistically significant (F5,30 = 1.09, P = 0.38). Females tended to have greater
concentrations of compound (1) than hermaphrodites, except in the CV population (Figure 2), when samples from this population were excluded from the
analysis, differences between genders were statistically significant (F1,24 = 5.76,
P = 0.025). Although the patterns were similar, gender differences were less
apparent for concentrations of compounds (2) and (3) (analyses not shown). Populations also differed in the average concentration of compounds (2) and (3). Plants
of the CT population had the highest concentration of both coumarin glycosides
(14,471.0 ± 3986.0 ppm and 22,763.8 ± 7203.7 ppm for (2) and (3), respectively), whereas plants of the RH population showed the lowest concentrations of
both (7614.9 ± 1818.4 ppm, and 14,903.7 ± 2355.6 ppm, respectively).
Since the concentration of the three coumarin glycosides was positively correlated, we calculated total concentration of coumarins in D. laureola leaves by
adding them. The average concentration of total coumarins in female individuals
of different populations was positively correlated to site altitude (N = 6, r = 0.90,
P = 0.01; Figure 3). However, such a relationship was not found for hermaphrodite
individuals (N = 6, r = 0.27, P = 0.60; Figure 3).
DISCUSSION
The three most abundant coumarin glycosides found in leaves of D. laureola were previously reported from other natural sources. Strangely, compounds
(1) and (2) were isolated from mosses (Jung et al., 1994), compound (1) was
found in Polytricum formossum and Atrichum undulatum, and compound (2) only
in P. formossum. Compound (3) was previously isolated from leaves of the congeneric Daphne pseudo-mezereum (Konishi et al., 1993). No evidence for a similar
coumarin daphnetin (7,8-dihydroxy coumarin) was detected in D. laureola leaves,
despite a former report on its presence in the bark of this species (cf. Murray et al.,
1982). This highlights that further analyses are needed to determine the identity
and abundance of coumarins in other plant structures. The three compounds found
in leaves are 5,7,8 trioxygenated coumarins. Apparently, all coumarins with an
oxygen-containing substituent at the 7-position seem to be biosynthetically distinct from those that lack such a function, and derived from p-coumaric acid, an
intermediate in lignin biosynthesis (Brown, 1970). It is also remarkable that all D.
laureola coumarins share a 7-methoxy function, differing only in the nature and
GENDER DIMORPHISM AND ALTITUDINAL VARIATION OF SECONDARY COMPOUNDS
145
TABLE 1. CHEMICAL STRUCTURE AND 1 H (J HZ) AND 13 C NMR DATA (500 MHZ,
DMSO-d6 ) OF THE THREE COUMARINS OBTAINED FROM THE METHANOLIC EXTRACT
OF Daphne laureola LEAVES, AND THE REFERENCES WHERE THESE COMPOUNDS WERE
PREVIOUSLY REPORTED
Compound
1
2
3
1H
(J Hz) and
R1
R2
O-β-1-glc
O-β-1-glc-6←1-β-glc
H
H
13 C
2
3
4
5
6
7
8
9
10
Ome
Glucose
1’
2’
3’
4’
5’
6’
1”
2”
3”
4”
5”
6”
Jung et al. (1994)
Jung et al. (1994)
Konishi et al. (1993)
O-β-1-glc
NMR data
1
Aglycone
References
1H
2
13 C
3.85 s
159.8
110.1
139.4
146.1
96.5
150.8
127.7
142.2
103.5
55.8
4.80 m
3.29 m
3.30 m
3.12 m
3.37 m
3.39 m/3.75 m
102.2
73.2
76.2
70.0
77.4
60.8
6.21 d (9,7)
8.23 d (9,7)
6.91 s
1H
3
13 C
3.85 s
161.4
111.9
140.7
147.6
97.9
152.4
–
143.8
106.6
57.3
4.86 d (7.0)
–
–
–
–
–
4.17 d (7.6)
3.03 t (10.9)
3.11 t (8.5)
–
2.95 t (7.9)
–
102.9
74.1
77.1
70.7
76.7
70.0
105.3
74.3
77.6
70.8
74.3
66.6
6.24 d (9.8)
8.25 d (9.6)
6.85 s
1H
13 C
3.84 s
160.3
109.4
139.8
151.9
95.6
155.8
124.1
147.9
102.5
56.2
4.93 d (7.3)
3.29 t (7.9)
3.24 t (8.4)
3.17 t (8.4)
3.08 m
3.61 m/3.37 m
102.4
74.1
76.5
69.9
77.2
60.9
6.11 d (9.6)
8.04 d (9.7)
6.52 s
146
ALONSO ET AL.
FIG. 2. Average concentration (+SE) of the most abundant coumarin, 5-O-β−D-glucosyl7-methoxy-8-hydroxy coumarin, in leaves of female and hermaphrodite D. laureola individuals from six populations in southern Spain. Note that populations are ordered from
lower to higher elevation.
position of the glycoside substituent, suggesting a common biosynthetical pathway. Further studies are needed to elucidate the metabolic relationships between
these compounds and other plant physiological processes such as leaf maturation,
since it is known that hydroxycoumarin content in Daphne mezereum from the
Moscow region is maximal during leaf formation and at the end of the growth
period (cf. Murray et al., 1982).
Our estimates of total coumarin concentration averaged between 60 and
120 mg/g dry weight for mature summer leaves of D. laureola. This figure is
ca. 10 times higher than the concentration of dihydroxycoumarins reported for
mature summer leaves of Daphne mezereum (Zobel and Brown, 1988), although
the difference could partially reflect differences in the accuracy of the methods
applied. The use of esculetin as internal standard in all samples allowed us to
estimate the actual process errors and be confident of our estimates. Moreover, the
low rates of standard recovery we observed revealed that further efforts to improve
our methods would be important it to detect less abundant compounds.
Female and hermaphrodite individuals of D. laureola growing in the same
population can have differential concentrations of secondary compounds. As
GENDER DIMORPHISM AND ALTITUDINAL VARIATION OF SECONDARY COMPOUNDS
147
FIG. 3. Average concentration of the sum of all coumarins found in hermaphrodite and
female D. laureola individuals at six different Spanish populations and the relationship
with site altitude (solid line for females and dashed line for hermaphrodites).
expected from optimal theory, due to lower allocation to flowers, female leaves
had on average higher concentrations of coumarins, upholding the idea that male
function is costly for hermaphrodites (Eckhart and Seger, 1999). However, the
magnitude of the difference was not constant, and even in the CV population, the
one at the lowest altitude and with the highest proportion of females, the sign of
the difference was reversed (Figure 2). Thus, costs of male (and likely also female)
reproduction seem to be context dependent.
Finally, we found that average coumarin concentrations differed among D.
laureola populations within a relatively small region. Heterogeneous spatial distribution of allelochemicals seems to be ubiquitous in both managed and natural
systems (Hoy et al., 1998). Occurrence and concentration of plant allelochemicals may vary with latitude, elevation, sun exposure, and other environmental
factors (Louda and Rodman, 1983; Dudt and Shure, 1994; Johnson and Scriber,
1994; Salmore and Hunter, 2001; Gómez et al., 2003). A negative relationship
between elevation, occurrence, and concentration of some alkaloids (Salmore and
Hunter, 2001) and glucosinolates (Louda and Rodman, 1983) has been found in
some species, although unrelatedness and nonlinear relationships were found for
148
ALONSO ET AL.
different alkaloids and glucosinolates in the same species. Also the content of
UV-B absorbing compounds increases with site elevation in several plant species
(Rozema et al., 1997). D. laureola showed a gender-specific response to site altitude
since concentration of coumarins in leaves of female shrubs increased with population altitude, but hermaphrodites did not show a similar altitudinal pattern. On
one hand, a higher concentration of coumarins could benefit plants by increasing
the UV absorbance in higher elevation sites. Apparently, mostly females would be
able to benefit from this advantage, once more supporting the existence of a cost of
the male function in this species. Interestingly, some field experiments have shown
that plants exposed to ambient solar UV-B radiation are more resistant to herbivorous insects than plants grown under filters that excluded the UV-B component
of solar radiation (Stratmann, 2003). In addition to the flavonoids, isoflavonoids,
and tannins quoted by Stratmann (2003), coumarins could be also associated with
the observed overlapping between plant physiological responses to UV radiation
and herbivory. Ongoing studies aimed at specifically evaluating the defensive role
of D. laureola coumarins against insect defoliation will help to clarify the consequences of gender and altitudinal variation in coumarin concentrations in leaves
herein reported.
Acknowledgments—We are particularly indebted to Carlos M. Herrera, Manuel M. Lomas, Jesús
Méndez, and Adelina Vázquez for advice and support. Special thanks to Isabel M. Garcı́a and Asunción
Vargas, who assisted in the lab, and to Tenji Konishi for providing coumarins. We also thank Consejerı́a de Medio Ambiente Junta de Andalucı́a for authorizing us to work in Cazorla; and the Estación
Biológica de Doñana, the Consejerı́a de Educación y Ciencia Junta de Andalucı́a, and the Spanish
Ministerio de Ciencia y Tecnologı́a (MCyT, BOS2003-02235) for funding the study. C.A. gratefully
acknowledges a grant from the Ramón y Cajal Program from MCyT. The manuscript was improved
thanks to comments of two anonymous referees and English revision by Pauline Agnew.
REFERENCES
ÅGREN, J., DANELL, K., ELMQVIST, T., ERICSON, L., and HJÄLTÉN, J. 1999. Sexual dimorphism and
biotic interactions, pp. 217–246, in M. A. Geber, T. E. Dawson, and L. F. Delph (eds.). Gender
and Sexual Dimorphism in Flowering Plants. Springer, Berlin.
ALONSO, C. and HERRERA, C. M. 1996. Variation in herbivory within and among plants of Daphne
laureola (Thymelaeaceae): Correlation with plant size and architecture. J. Ecol. 84:495–
502.
ALONSO, C. and HERRERA, C. M. 2000. Seasonal variation in leaf characteristics and food selection
by larval noctuids on an evergreen Mediterranean shrub. Acta Oecol. 21:257–265.
ALONSO, C. and HERRERA, C. M. 2001. Neither vegetative nor reproductive advantages account
for high frequency of male-steriles in southern Spanish gynodioecious Daphne laureola
(Thymelaeaceae). Am. J. Bot. 88:1016–1024.
ALONSO, C. and HERRERA, C. M. 2003. Developmental and spatial covariation of nutrients in growing
leaves of Daphne laureola and their relationships with herbivory. New Phytol. 159:645–656.
GENDER DIMORPHISM AND ALTITUDINAL VARIATION OF SECONDARY COMPOUNDS
149
ASHMAN, T.-L. 2002. The role of herbivores in the evolution of separate sexes from hermaphroditism.
Ecology 83:1175–1184.
BERENBAUM, M. R. 2001. Chemical mediation of coevolution: phylogenetic evidence for Apiaceae
and associates. Ann. Mo. Bot. Gard. 88:45–59.
BOECKLEN, W. J., PRICE, P. W., and MOPPER, S. 1990. Sex and drugs and herbivores: sex-biased
herbivory in arroyo willow (Salix lasiolepis). Ecology 71:581–588.
BROWN, S. A. 1970. Biosynthesis of furanocoumarins in parsnips. Phytochemistry 9:2471–2475.
CIPOLLINI, D., PURRINGTON, C. B., and BERGELSON, J. 2003. Costs of induced responses in plants.
Basic Appl. Ecol. 4:79–85.
COLLIN, C. L., PENNINGS, P. S., RUEFFLER, C., WIDMER, S., and SHYKOFF, J. A. 2002. Natural enemies
and sex: How seed predators and pathogens contribute to sex-differential reproductive success in
a gynodioecious plant. Oecologia 131:94–102.
DELPH, L. F. 1990. Sex-differential resource allocation patterns in the subdioecious shrub Hebe
subalpina. Ecology 71:1342–1351.
DUDT, J. F. and SHURE, D. J. 1994. The influence of light and nutrients on foliar phenolics and insect
herbivory. Ecology 75:86–98.
ECKHART, V. M. and SEGER, J. 1999. Phenological and developmental costs of male sex function in
hermaphroditic plants, pp. 195–213, in T. O. Vuorisalo and P. K. Mutikainen (eds.). Life History
Evolution in Plants. Kluwer, Dordrecht.
GÓMEZ, D., AZORı́N, J., BASTIDA, J., VILADOMAT, F., and CODINA, C. 2003. Seasonal and spatial
variations of alkaloids in Merendera montana in relation to chemical defense and phenology. J.
Chem. Ecol. 29:1117–1126.
GOUYON, P. H. and VERNET, P. 1980. Étude la de variabilité génétique dans une population de
Thymus vulgaris L. Observations sur le polymorphismes sexuel et chimique, effets du régime de
reproduction. Acta Oecol., Oecol. Plant. 1:165–178.
HEGNAUER, R. 1973. Chemotaxonomie der Pflanzen. VI. Thymeleaeceae. Birkauser Verlag, Basel.
HERMS, D. A. and MATTSON, W. J. 1992. The dilemma of plants to grow or to defend. Quart. Rev.
Biol. 67:283–334.
HOY, C. W., HEAD, G. P., and HALL, F. R. 1998. Spatial heterogeneity and insect adaptation to toxins.
Annu. Rev. Entomol. 43:571–594.
JING, S. W. and COLEY, P. D. 1990. Dioecy and herbivory: The effect of growth rate on plant defense
in Acer negundo. Oikos 58:369–377.
JOHNSON, K. and SCRIBER, J. M. 1994. Geographic variation in plant allelochemicals of significance to insect herbivores, pp. 7–31, in T. N. Ananthakrishnan (ed.). Functional Dynamics of
Phytophagous Insects. Science Publishers, Lebanon, New Hampshire, USA.
JUNG, M., ZINSMEISTER, H. D., and GEIGER, H. 1994. New three- and tetraoxygenated coumarin
glucosides from the mosses Atrichum undulatum and Polytrichum formosum. Z. Naturforsch 49
c:697–702.
KONISHI, T., WADA, S., and KIYOSAWA, S. 1993. Constituents of the leaves of Daphne pseudomezereum. Yakugaku Zasshi J. Pharm. Soc. Jap. 113:670–675.
LOUDA, S. M. and RODMAN, J. E. 1983. Ecological patterns in the glucosinolate content of a native
mustard, Cardamine cordifolia, in the Rocky Mountains. J. Chem. Ecol. 9:397–422.
MARSHALL, M. and GANDERS, F. R. 2001. Sex-biased seed predation and the maintenance of females
in a gynodioecious plant. Am. J. Bot. 88:1437–1443.
MAYNARD-SMITH, J. 1978. Optimization theory in evolution. Annu. Rev. Entomol. 9:31–56.
MURRAY, R. D. H., MÉNDEZ, J., and BROWN, S. A. 1982. The Natural Coumarins. Occurrence,
Chemistry and Biochemistry. Wiley, Chichester, UK.
OBESO, J. R. 2002. Tansley review no. 139. The costs of reproduction in plants. New Phytol.
155:321–348.
150
ALONSO ET AL.
ROZEMA, J., CHARDONNENS, A., TOSSERAMS, M., HAFKENSCHEID, R. and BRUIJNZEEL, S. 1997. Leaf
thickness and UV-B absorbing pigments of plants in relation to an elevational gradient along the
Blue Mountains, Jamaica. Plant Ecology 128:151–159.
SALMORE, A. K. and HUNTER, M. D. 2001. Elevational trends in defense chemistry, vegetation, and
reproduction in Sanguinaria canadensis. J. Chem. Ecol. 27:1713–1727.
SAS Institute. 1996. SAS/STAT Software: Changes and Enhancements Through Release 6.11. SAS
Institute, Cary, North Carolina.
STRATMANN, J. 2003. Ultraviolet-B radiation co-opts defense signalling pathways. Trends. Plant. Sci.
8:526–533.
THOMPSON, H. J. and BROWN, S. A. 1984. Separations of some coumarins of higher plants by liquid
chromatography. J. Chrom. 314:323–336.
ULUBELEN, A., TEREM, B., and TUZLACI, E. 1986. Coumarins and flavonoids from Daphne gnidioides.
J. Nat. Prod.49:692–694.
UNO, G. E. 1982. Comparative reproductive biology of hermaphroditic and male-sterile Iris
douglasiana Herb. (Iridaceae). Am. J. Bot. 69:818–823.
ZOBEL, A. M. and BROWN, S. A. 1988. Localization of daphnetin and umbelliferone in different
tissues of Daphne mezereum shoots. Can. J. Bot. 67:1456–1459.